[00:25] hatch: still around? [01:00] rick_h_: back [01:06] hatch: nvm, thanks. I'm still getting multi-dispatch or something but some of it is my own fault. [01:07] hatch: and I wasn't sure if your multi-dispatch killer landed for sure [01:11] rick_h_: ohh - well it only killed one instance that was causing the multi dispatch [01:11] :) [01:12] hatch: ah ok. Then I won't raise a flag then [01:12] current branch kills another....but still doesn't kill the primary culprit [01:12] hatch: cool, thought it might be done/gone. [01:13] heh - it's pretty well integrated into the app so it will probably be a while [01:13] I have a branch which fixes it....albeit very hackily and the app is so fast haha [01:14] yea, I was thinking today you had mentioned a speed improvement and I was thinking "man, I don't notice it" [01:14] your computer is also 2x as fast as mine though :P [01:16] heh, but still I should get faster too [01:16] rick_h_: Any ideas why the gui might not be loading on a fresh branch today? It hangs on "Connecting to the Juju environment" and it's not spitting out any errors. [01:17] huwshimi: only time it does that is if I don't have the rapi runningin the background [01:17] ugh [01:17] rick_h_: Nevermind :) [01:17] :) [01:17] been there done that [01:18] :) [01:18] haha [01:18] yeah it should really have a timeout error [01:18] :) [01:18] at least make devel should just dump out "hey dude...you're missing a step" [01:19] or even just start it for you [01:19] I've got a make run command that starts my webapp, a combo loader, and celery so you can use/test things [01:19] ahh that's cool [01:19] I have a node server installed called `lactate` so I type lactate in any dir and it's a webserver [01:19] works great for dev [01:20] https://npmjs.org/package/lactate [01:21] hatch: http://www.linuxjournal.com/content/tech-tip-really-simple-http-server-python :) [01:21] no extra install req'd :P [01:21] ahh cool - I actually prefer node's limited core install :) [01:22] but I should have guessed python would have something similar [01:24] ok, I can sleep at night now. Figured out wtf was going on. [01:25] and yay, NAS backup to usb drive is 29/36 hrs done [01:26] haha wow large NAS [01:26] 1.2TB out of the 2TB. Going to upgrade it to 3TB disks and take the 2TB out and put them in the desktop later on. [01:27] but it's only usb2 and an atom model so taking FOREVER to back it up before I can swap the disks and grow the raid [01:28] ahh yeah that would be slow [01:28] :) [01:28] oh well, only have to do it once in a blue moon [03:06] yeah true true [03:08] YUI 3.10 will likely go live tomorrow so we should investigate upgrading [04:45] gary_poster: thanks, sorry I missed your mention earlier. Yeah I wanted to make sure this worked first before moving it to the db code [10:51] * benji rubs the sleep from his eyes. [10:59] benji: +1 [10:59] jcsackett: ping when you get in. Think I messed up :/ [10:59] :) [11:27] gary_poster: when you say we should be looking at them on OSX do we have a system for that? /me has no apple in the house. [12:07] rick_h_, I have a mac I can dual boot into. Also, if it is simply a missing font because the browser is not in Ubuntu, we should be able to see that on IE 10, which we want to test on anyway [12:08] rick_h_, do you have a win machine or vm? [12:18] gary_poster: yea, I have a win vm. I'll admit I probably wouldn't notice the font in that section itself. [12:18] :-) [12:31] bac I looked at jovan's suggestion to see how G+ did it. Did you look at that already? [12:31] i just noticed that when running the test-server the URL produced by clicking on a 'describes' heading is now correct and doesn't have to be manually edited. big thumbs up to whoever fixed that. [12:31] gary_poster: no, i was trying to wrap up my other branch [12:32] makes sense bac. I'll play with it for a sec and see what I find. They are using a div with content editable, not a textarea. [12:32] bac: heh, url was correct but the file wasn't loaded. That was hatch and gary_poster :) [12:33] rick_h_: i will buy them a beverage of their choice in oakland [12:33] :-) [12:33] preferably at breakfast [12:34] offer limited to hotel-provided breakfast. not combinable with other offers. some restrictions apply. not valid in nevada or california. [12:34] :-P [12:35] "here's your bagel, don't ask for any cream cheese...enjoy" [12:35] heh [12:36] i do hope the hotel has an omelet station. it's the only thing that makes work travel bearable. [12:49] bac, http://paste.ubuntu.com/5598238/ is a mostly minimal approach that seems to have the behavior we want. Did not know how to do that so wanted to figure it out [12:49] Nice that it is pure css [12:49] luca__: morning, do you guys have an icon for the second 'state' of the fullscreen/sidebar toggle button? Huw's asking for it. [12:50] bac, hmm. does not seem to work in ff :-/ [12:50] gary_poster: cool. will look in a bit [12:51] gary_poster: http://caniuse.com/#feat=contenteditable should? [12:51] rick_h_: We do, I'm checking with Greg now to see if he can find/send you the asset. [12:51] rick_h_, contenteditable does, not that specific page :-) [12:51] luca__: appreciate it. see bug https://bugs.launchpad.net/juju-gui/+bug/1171533 [12:51] <_mup_> Bug #1171533: There is no "minimised" icon state for the state of canvas icon

[12:51] gary_poster: ah, gotcha. [12:56] rick_h_: Greg has pointed me towards this folder: https://drive.google.com/a/canonical.com/?usp=chrome_app#folders/0B7yNWRv_QU7WVnF2MGktY00wUEE [12:56] works in FF and Chrome: http://paste.ubuntu.com/5598256/ [12:57] trying ie [12:57] rick_h_: however its quite clear that he hasn't completed all of the assets, I'll raise this as a priority task and get them out to you guys asap. [12:58] luca__: ah cool. I don't recall seeing this folder in there before. Was this recently shared? [12:58] luca__: ok, well I'll pass this along to Huw and yea, the ones he's asking for are missing to help complete the bug brought up. [13:53] mornin [13:53] gary_poster: we're co-opting our hangout for the ux call https://plus.google.com/hangouts/_/cf491220dfb7736cee2876baf29110eeb52f5997 [13:53] morn hatch [13:57] hatch: since you just redid a bunch of work around interfaces. Are the peer relations used in the gui? I see some code in the environs, but not actually hitting model.get('peers')? [13:58] peer relations are not used in the gui right now [13:58] hatch: awesome [13:58] agreed - i hate relations lol [13:59] rick_h_, ack, thx [13:59] rick_h_, hatch, we should support peers, so please don't make that harder :-) [13:59] gary_poster: ok, well right now the backend (charmworld) doesn't present the info in the api so we were looking at if it needs to be implemented/not atm [14:00] gary_poster: so not 'harder' but 'delayed'? [14:00] noooooooooo [14:01] I'll be sick the day we need to implement peer relations mkay? ;) [14:01] heh ok [14:01] haha [14:02] last night I went to a flooring auction [14:03] came hom ewith nothing :( [14:06] flooring auction eh? big night at the hatch household [14:07] hah - well we decided that we would renovate instead of move so over the next year I'm trying to collect all the supplies :) [14:09] I figure if I do it right I can probably save 30% [14:11] anyone itching to do a review? https://codereview.appspot.com/8935043 [14:12] * hatch raises hand [14:12] hatch: you practicing for the next auction? [14:13] I think to practice for that I'll need to warm my wallet up [14:13] the prices were nuts [14:13] did every item end with "sold to the guy in flannel?" [14:15] a number of them did.... [14:15] * hatch wonders if bac is actually a Saskatchewanian [14:17] i just can't envision a flooring auction in saskatchewan playing out any differently [14:18] there was one guy in a suit [14:18] possibly a flooring mobster [14:18] one of the hunt brothers, cornering the market no doubt [14:19] hatch: yep, you should be able to save 30% over having the work done, if you're super frugal you can probably get to 50%. I do a lot of my own work and would be glad to help out however I can. [14:21] * hatch flys benji here to work [14:21] :) [14:21] bac: review done [14:22] thanks [14:23] :) [14:35] any rap/hiphop fans? [14:35] how about electronica? [14:38] Electronica? [14:39] trance, techno, house, dubstep [14:39] Heh, I spend most evenings getting my ear talked off (in a good way) about all the different genres that fall under that umbrella, but I've not really heard the word in forever. [14:40] hatch: I like some of those, but not enough to purchase much or know many of the artists [14:40] Makyo: heh yeah there are a lot so I usually start with electronica then drill down from there :) [14:41] http://techno.org/electronic-music-guide/music.swf [14:41] haha di.fm, I have a subscription to there [14:41] benji, , yeah, that's pretty awesome. Showed that to my professor and we talked about that in studio for two weeks. [14:41] benji: I forgot about this swf :) thanks....bookmarked [14:42] I pretty much listen to trance all day long [14:42] oldie but a goodie [14:43] 140bpm is good for coding :P [14:43] Trance > Dutch [14:43] Hah [14:44] lol they have some funny names on here [14:44] buttrock goa [14:45] and the comment is hilarious [14:45] Ishkur's got some opinions :) [14:46] it needs to be updated! [14:46] it's missing dubstep! [14:48] gary_poster: did your little html test actually do word breaks in chromium? [14:48] Do it in D3. [14:48] With HTML5 audio [14:48] gary_poster: nm [14:48] who has that kind of time? [14:48] bac http://jsfiddle.net/nzHKE/ was hacking this [14:48] maybe a slack task? ;) [14:48] bac on call [14:49] Haha [14:51] Makyo: actually it could probably be done in a weekend with a couple guys [14:52] hatch, digraphs are pretty easy with some of the layout, and now there's http://trifacta.github.io/vega/ which could make it faster. [14:52] s/layout/layouts [14:52] ahh cool cool [14:53] reviews and qa plz https://codereview.appspot.com/8837049/ [14:55] hatch: so, i like things named dubstep, but i'm still trying to figure out what really distinguishes it as anything more than a subtype of DnB. [14:57] well imho there is real dubstep and then skrillex dubstep [14:57] :) [14:57] Brostep. [14:57] yeah that [14:57] hahaha [14:57] hatch: so you've introduced yet more subcategories, but still no real distinction from DnB. :-P [14:58] And then there's chipstep: http://chibitech.bandcamp.com/track/moe-moe-kyunstep-part-ii [14:58] More impressive for being done entirely on NES hardware :P [14:59] jcsackett: everything in electronica is a subcategory :) [14:59] that's fair. :-P [14:59] Makyo: HAHAHAHAHA [15:00] Makyo: interesting. i do like chiptunes... [15:00] jcsackett, Me too :) This is one of my favorites recently: https://soundcloud.com/henryhomesweet/sets/enter-5d [15:01] and even with all this talk I'm listening to Def Leopard right now haha [15:02] jcsackett, Henry Homesweet's pretty neat, since his music's all hardware, most of the time. http://www.youtube.com/watch?v=CV2hslJjs9U [15:02] oh nice. [15:02] this is pretty chill [15:02] so, actual chips in the chiptunes. :-P [15:04] Heh, yeah :) [15:05] current podcasts here Above&Beyond Group Therapy, Corstens Countdown, Hardwell On Air, Paul van Dyk's VONYC Sessions, The Sound Of Trance, A State Of Trance, Thrillseekers Nightmusic Podcast [15:05] * Makyo adds https://itunes.apple.com/us/podcast/turn-it-around-electronic/id563798064 to that list. [15:06] found this cool track last night https://itunes.apple.com/ca/album/lose-sight-feat.-ane-brun/id624001460?i=624001717 [15:07] thanks for the podcast, I'll add it to the list :) [15:07] hatch, Whoa, I was just pointed at that yesterday. Good stuff. [15:08] haha yeah? I am a fan of Andrew Bayer [15:08] Might like http://www.youtube.com/watch?v=ADBKdSCbmiM too, hatch. [15:08] If you like Royksopp. [15:08] Though it's mostly similar in the video. [15:09] ahh yeah great song [15:09] messed up video [15:09] Hah, yeah, [15:11] bac did you see how that worked and didn't work? I didn't get the extraction to work the way we wanted [15:11] that will be the annoying bit of this I think [15:11] Because people can also paste in arbitrary stuff too! [15:11] gary_poster: i haven't had much success with it. does the box auto grow in height for you? [15:11] bac, yes [15:12] hm, doesn't for me in chromium on quantal [15:13] gary_poster: oh, it does with new lines...which is what we want. duh. i was expecting auto-wrap like the g+ example [15:13] right [15:13] Makyo: this is a really nice set. [15:18] jcsackett, yeah, I like a lot of his stuff. [15:21] jcsackett: can you peek at https://codereview.appspot.com/8593050/ please? [15:21] rick_h_: looking. [15:22] https://codereview.appspot.com/8837049/ [15:22] gary_poster: care to look at https://codereview.appspot.com/8935043 ? should be quick. [15:23] wil try bac. on call [15:23] thanks jcsackett [15:23] rick_h_: lgtm. [15:23] hatch: on it [15:23] thanks [15:23] qa if you could please [15:23] :) [15:24] I'm always scared of a net negative diff [15:24] :) [15:24] those are the best kinds! [15:31] frankban: yay! Thanks for working on that login bug :) [15:32] hatch: almost ready for review [15:32] oh right on u rock [15:33] hatch: so you ripped out all of the prefetch service bits. that was redundant now that handleServiceEvent is doing it when a service is added, no? [15:33] bac: correct - as soon as the service is added it pulls in the extra data [15:33] so it's always available when needed [15:33] endpoint.js is becoming a worse name by the day [15:34] well all of that code will be moved into the db, likely this week [15:34] it was just the logical place to put it in a multi step process [15:51] jujugui please update kanban [15:52] gary_poster: I've got another call and will be out. [15:52] ack rick_h_ thanks for heads up [15:58] jujugui call in 2 [15:58] frankban: review done, running make for QA [15:58] hatch: cool, thanks [16:00] benji call now [16:10] hatch: neither, san juan [16:11] ahh ok I was just looking at expedia and those are the only two airports it had :) [16:11] hatch: SJU [16:11] hatch: your branch is LGTM pending QA [16:12] hum... charm install-error in CI [16:13] ahh I missed SJU === _mup__ is now known as _mup_ [16:25] yowzas flights are expensive to SJU [16:27] it varies wildly by season. my flight to SFO was pretty cheap [16:27] guihelp: so juju-core doesn't have debug-hooks. is there a work around? what does one do when the service doesn't start? [16:30] bac, log into machine (not with juju) examine log files [16:30] bac, debug-log is present although it doesn't properly qualify all output by unit [16:30] bac, and then juju resolved --retry unit_name [16:31] to have it go again [16:31] frankban: your branch is way more responsive thanks for that fix [16:31] hazmat: ok. i just liked the ability to run the hooks interactively [16:33] frankban: if I visit http://192.168.2.119:8888/:gui:/service/memcached/ on `make prod` after logging out, I only get a blank crosshatch [16:33] well substitute that ip for localhost [16:33] :) [16:33] hatch: is this different in trunk? AFAICT make prod does not support fallback resources [16:34] I'll check trunk [16:34] thnaks [16:34] thanks even [16:36] ok same on trunk [16:39] hatch: 3.10 is out!! [16:40] hatch: speed speed speed! [16:40] hatch: ruh roh: you’ll need to change your code to use the new static modifyAttrs() method. [16:44] rick_h_: I want to see how fast the tests run using 3.9.1 vs 3.10 to see if the speed differences make any real world impact [16:44] hatch: well I'll let you do it then to see the biggest diff :P [16:44] haha [16:45] bcsaller: db means db.service yes [16:46] hatch: thanks [16:46] the idea is that when you request a service from the db it will always return a promise with a fully populated service [16:46] no more half data garbage [16:46] :) [16:50] agreed [16:52] hatch: i was trying to qa your branch on ec2 but the juju-gui charm is failing. [16:52] not your branch's fault [16:53] bac: oh you can do it locally with rapi no? [16:54] hatch: yeah, i did that and it looked good. just thought i'd be more paranoid. i'll approve it now. [16:54] haha well thanks - if there are any errors I'm sure they will come up soon, although the code will be gone very soon [16:55] Makyo and hatch: can you review https://codereview.appspot.com/8940043 for rick_h_ and me? [16:55] jcsackett, Sure thing. [16:55] I can yep [16:56] thanks! [16:58] jcsackett: I am pretty sure that we decided for /** and */ with no *'s [17:03] hatch: that's *wildly* inconsistent throughout the code then, and the majority of code i have seen has been with * [17:04] jcsackett: yea, recent style decision [17:04] jcsackett: that's my fault. I still mix it up sometimes and didn't really share that back with you. [17:04] it is inconsistent but you went through only one file in your diff and changed it...making it even more inconsistent lol [17:04] hatch: it's the only file i've seen it in without recently. [17:04] since within the subapp it's almost entirely * [17:05] also: i *believe* the linter notices commenting errors only with *. after i added those in, it yelled about a param comment not ending with a period, but skipped it before. [17:05] jcsackett: yea, it's just because no one wants to do the big diff to change them all [17:06] jcsackett: but it's the agreed standard going forward and we're trying to retrain that way [17:06] rick_h_: i mean, i suppose i can uncommit that revision. [17:06] why the recent decision? [17:06] jcsackett: it was brought up on a friday call and agreed to. [17:07] sure, why? what's the argument for/against? [17:07] because maintaining the *'s is a pita for some. [17:07] fair. [17:07] and readability can be better without the * I think [17:07] i disagree, actually. :-P [17:08] but if i'm alone on that, i don't see a need to fight it. [17:08] heh, I tried :) [17:08] but it was a fiar '+1, 0, -1' vote and the *'s lost :) [17:09] damn democracy [17:09] it was really the ones who abstained who are causing the issues [17:09] :P [17:10] lol [17:10] well it's my fault. I didn't make sure to tell jcsackett sorry :( [17:10] frankban: now that you have 2x reviews can you merge with trunk so that I can merge it into my current branch :) [17:10] hatch: I am on it [17:10] thanks sir [17:10] hatch: and thanks for your review [17:11] * benji lunches/travels to work site number 2. [17:11] cd .. [17:11] oops [17:12] at least it wasn't an admin password [17:12] :) [17:12] jujugui do we know why CI is failing with install-error? Or should someone investigate [17:14] It ran fine yesterday, is it possible the npm ppa change broke it? [17:14] ok, just booting up the laptop to log into it [17:14] hatch we don't know why, and I just asked wedgwood to merge it. yeah, teknico escaped just in time :-) [17:14] haha [17:15] man I really wish the console showed *'s for passwords [17:15] I have no idea what window I'm typing them into half the time lol [17:19] hatch: landed [17:19] thanks [17:20] have a great week all! [17:20] you too frankban! have fun! === deryck is now known as deryck[afk] [17:27] hatch, discover anything? bcsaller, I tried to deploy juju-gui in pyJuju and got this: http://pastebin.ubuntu.com/5598870/. Is that indicative of a problem in the charm store? [17:27] gary_poster: it restarted so I need to wait for it to fail agian [17:27] ok hatch [17:28] that definitely looks like a charm store error to me [17:28] I've never seen it before [17:28] I'll get Ben's opinion before throwing it to orange squad (or whoever looks at this stuff?) [17:30] write it on a piece of paper and slide it under their door then quickly walk away [17:30] ;) [17:30] :-) [17:32] hatch do you have pyjuju installed? [17:32] rapi? [17:32] no hatch, the real one, installed from pyjuju PPA. [17:33] sinzui, I'm getting http://pastebin.ubuntu.com/5598870/ when I try to deploy in pyjuju on quantal. Do you happen to know if that is a charm store issue? I'm trying to see if I can get someone to dupe [17:33] umm I'm not sure let me see [17:33] bac you around, and do you still have a pyjuju you can use?? [17:33] doesn't look like it [17:33] (sorry, only one question mark intended) [17:33] I can install it [17:33] s'ok hatch, thanks [17:34] alrighty [17:34] hatch, unless I can't get anyone else to look at it ;-) then you are on the hook [17:34] haha booo [17:34] gary_poster, I haven't seen that. I am on raring [17:34] * sinzui runs the command [17:34] sinzui, I have the go version on raring and they appear to be incompatible :-/ [17:35] they both want to own /usr/bin/juju [17:35] ah [17:39] sinzui, you don't dupe? [17:40] gary_poster, no. deployed without error [17:40] and the service is started [17:40] sinzui, ok thanks. digging locally === matsubara is now known as matsubara-afk [17:44] hmm I can't ssh into the box [17:44] it keeps timing out :/ [17:47] now I'm not even sure how to debug it [17:48] maybe trigger it again and hope that the debug-log catches it [17:48] well [17:48] we should be able to deploy it within our own local juju [17:48] and dupe [17:48] if it is a problem with the charm itself [17:49] ok is there hacking docs for that? I don't think I've ever done that [17:49] hatch for pyjuju. https://juju.ubuntu.com/docs/getting-started.html [17:51] oh - yeah I have that up and running, but only for EC2 [17:51] haven't done it with LXC, sorry I misunderstood [17:52] hatch, yeah EC2 is fine for this. I discovered that I was running into https://bugs.launchpad.net/juju/+bug/1050741 [17:52] <_mup_> Bug #1050741: juju bootstrap: ERROR [('PEM routines', 'PEM_read_bio', 'no start line')] [17:52] so easy work around [17:52] ahh ok - and you figure that's the same bug that the CI is running into? [17:52] or would you like me to go through the EC2 deployment [17:52] no hatch [17:52] I mean not the same bug [17:53] hatch I'll dig for a bit [17:53] ok lemme know if you want me to fire off the EC2 deployment [17:53] using juju 0.6.0.1 [17:55] hatch that's very old. You probably want to install the ppa as that doc describes [17:58] hatch charm works fine for me on ec2, at least with basic deploy :-/ [17:59] gary_poster: this on canonistack? abentley was seeing some issues on that today [17:59] rick_h_, yes. unable to install charm? [17:59] not sure, abentley any notes from the issues this morning? Or since staging is up now it just fixed itself? [18:00] gary_poster: oh sorry that's the one that is on the canonistack deployment [18:00] rick_h_, hatch: IS fixed canonistack and then staging came back. [18:01] rick_h_: It looks fine right now. [18:01] hatch what are are you apologizing for? :-) ok thanks abentley. how long ago was that fix, roughly? less than an hour ago? [18:01] canonistack fix, I mean [18:01] cuz I'm Canadian I guess :P [18:01] hatch :-) [18:02] I fired off another build [18:02] maybe it'll work with the fix that abentley mentioned [18:04] hatch: I mean that all of canonistack was broken, and then IS fixed all of canonistack. Right now, canonistack does not look broken to me. [18:09] lunching....I'll monitor this test as close as I can [18:42] yeah I think canonistack is broken again [18:42] the last 3 have failed with canonistack type errors [18:54] bac, fwiw http://jsfiddle.net/Pegsa/ . Fiddles with it while doing other things. Note dependency in "External resources." Kinda gross, but I like the UX effect, and I think everything else I'm doing is necessary. [18:54] but possibly not sufficient [18:54] gary_poster: i'll have a look in a sec [18:54] landing the second branch now [18:54] np, no rush [18:55] awesome [18:57] * benji remembers to start his IRC client. [18:58] benji: quiet day eh? [18:59] heh, well, for the last hour at least; I relocated during lunch === matsubara-afk is now known as matsubara [19:16] for very time-geeky reasons you shouldn't make a npm cache file on back to back seconds if one of those seconds is a leap second; I don't think that'll be much of a problem [19:17] wow, how did you think of that benji? :-) [19:18] lol! [19:18]